So what are the odds that Obama will be reelected in November despite the slow economic recovery?
Pretty darn good, according to these researchers’ analysis.
UCLA political scientist and Associate Professor Lynn Vavreck and John Sides, associate professor of political science at George Washington University, are confident that they can predict the outcome of the November vote, based on their study of 60 years of post-World War II U.S. presidential elections and other huge datasets. Since December, they have been analyzing data from public opinion polling done by YouGov, a polling firm in Palo Alto conducting 43,000 interviews online from a representative sample of people nationwide from Jan. 1 to Nov. 6, election day. Vavreck and Sides have also tracked news and social media content as well as campaign advertising to apply their social science perspective.
Vavreck and Sides call this their "Moneyball" approach — basing their data-intensive analysis of the election on statistical patterns and models, facts and reams of data, like the hero in the baseball movie. Rather than wait three years after the election to publish their book, as often happens in academic publishing, they’re bringing their research to readers in record time in order to influence the conversation that’s going on now among journalists and media commentators who will write the 2012 election into history.
This is the first time that political scientists are analyzing and writing an academic book in real time, Vavreck said. To assist them, their publisher, Princeton University Press, has compressed the time it takes to get it peer-reviewed.
As a result, a preview of the first two chapters of their book,"The Gamble: Choice and Chance in the 2012 Presidential Election," is now available for downloading free. But unlike a linear book, the authors plan to write the introduction after the election. The entire book will then be published in hardback and digital format by late summer next year.
During a brief break in her crammed schedule, Vavreck, author of "The Message Matters: The Economy and Presidential Campaigns," spoke with UCLA Today’s Cynthia Lee about their findings and the challenges of fast-tracking their book. The researchers also explain their decision in a video at the end of this story.
So let’s get to the bottom line first: What are President’s Obama’s chances of winning or losing if the election were held tomorrow? And couldn’t the predicted outcome in November be overturned by events that take place at any time — in a day or a week — during the campaign?
Obama is the likely winner. When we look back at the history of modern presidential elections over the last 60 years, we can tell you that the two most important factors that predict election outcomes are the state of the economy and party identification. And these fundamental considerations are not likely to change in a day or week. They are determined months before the election takes place.
There are certainly cases where the predicted the winner (based on these factors) loses. But they don’t lose it in a day or week. It happens in the beginning, in the messaging that’s crafted and carried out over the entire campaign. There are very few moments in American presidential campaigns that you can point to as THE moment that changed everything. It just doesn’t happen.
So you base your predictions solely on statistics, facts and data?
The statistics, facts and patterns help us understand the choices voters are making in this election. And that does translate into what we hope is the right prediction. But it’s really more about understanding the process. People always like to know why the winner won, so some journalists and historians invent stories. John F. Kennedy won because he was attractive, and Richard Nixon had flop-sweat. Dukakis lost because he got in a tank, made an ad and ground the gears. None of this really explains why people won or lost. So what we want to do is get the story about why the winner actually won tied to patterns, facts and data.
How does the economy help Barack Obama’s chances for reelection?
The economy is growing, but slowly. But even this slow amount of growth is enough historically to predict the reelection of the incumbent president. I think it is going to be a very close election. But being an incumbent gives him a bit of a boost. So he needs even less economic growth than, say, someone else from the Democratic Party who’s not an incumbent.
Our prediction is that Obama is likely to win. But that’s not a foregone conclusion. There are certainly things that Mitt Romney could do to "steal" the election from Barack Obama. But we don’t see him doing a lot of those things. So then we’re likely to say that the pattern will hold.
When you say the economy will help him get reelected, do you mean the state of the economy at that moment in November?
No, not at that moment. And it isn’t about your personal pocketbook situation either. It’s an assessment of the nation’s economic condition pretty much over the last six months or so, from January to the summer of the election year. For whatever reason, that is the predictor that helps us understand these aggregate election outcomes.
But Obama has been criticized for bailing out the banks and financial institutions and for the stagnant job market. Aren’t people blaming the financial crisis on him?
People are blaming George Bush for the country’s economic problems. You see this very clearly in the data. They seem to understand that Obama inherited a bad situation and that it all started with the Bush administration.
You said the other important factor that determines elections is party identification. Can you explain?
Obama’s approval rating has held very steady throughout his presidency — right there in the middle. That’s because Republicans hate him; they really hate him. And Democrats really love him. So people have just stuck to those opinions. Republicans are not going to move from their very low assessment of Barack Obama as president. And the Democrats are not going to move from their strong favorable assessment of Obama. Since that represents a good chunk of the country, that means his approval ratings have not waffled much at all.
In the past, strong Democrats might have thought Reagan did a good job and they would move a bit. Or strong Republicans might have actually approved of Clinton’s handling of his job. But now these sides are really well-formed. That makes it hard for there to be any crossover.
Compiling a statistical model of approval that took into account unemployment, the GDP and other factors, you compared that to Obama’s actual approval ratings and found that early in his presidency, Obama was slightly less popular than expected. But by the end of 2010 and continuing into 2012, he was more popular than expected. Why is that?
One of the interesting findings is that Obama is even more popular than he should be, given the state of the economy. So he’s just doing better with voters than we might expect, given these patterns.
And, again, it’s that level of commitment that people have to their parties. It actually works to help the president in this case.
The other reason we think his popularity is higher than expected is that there’s this general sense of likability that people have for him. He’s a nice guy, and people really do genuinely like him.
How has Romney’s selection of Paul Ryan as his vice president influenced the race?
Paul Ryan hasn’t done anything to change the state of play in this race. There’s no evidence that he’s helped the ticket. I think people really thought there would be a bounce in the polls for Romney on the heels of the Ryan nomination. And there has not been.
The line has held incredibly steady. The reason for that is fairly simple: Everybody sees the Ryan nomination as Romney’s attempt to shore up his base. The problem is Romney’s base has always liked him, and they still like him. His problem is with the middle, the moderates and the independents who might lean Republican. These are the people he needs to win over. They’re not sold on him yet. And Paul Ryan is not going to help move these people. So essentially, the group of people whom Ryan is popular with already like Romney an awful lot.
Despite what you read in the newspapers, it is not the case that the base is not enthusiastic about Romney. They are his most enthusiastic supporters. It’s the middle that isn’t sure what to do with him.
What are the challenges of fast-tracking the writing and publishing of an academic book?
We’re working with a lot of data. So the sheer number of man-hours it takes is part of the challenge. This book is coming out in real time, but it’s also peer-reviewed. So that’s a big challenge. We need to do the data analyses, write the chapters, and Princeton University Press then sends them out to our colleagues who will be critical, review them, find things we did wrong or tell us they are not convinced by our data.
All of this then comes back to us. We take account of these suggestions and make changes or don’t. Meanwhile, the book needs to be typeset, edited and we need to proof it again, plus deal with the graphics. … And we’re doing this ourselves. We have no staff.
But everyone has been very supportive. People have said, "You’re going to make mistakes." And, yes, we’re going to make mistakes. But when you find them, we’re going to fix them. You just have to take your ego out of it.
For us, the payoff is being a part of the conversation, and the journalists have been great about that. We went to Iowa, New Hampshire, traveled around and developed a relationship with a lot of journalists who’ve welcomed us to the conversation.
That payoff is worth the potential cost of someone discovering that one of our standard errors was wrong. So we’ll go fix that. Our egos can handle that.
Have you any sense whether your analyses have influenced the way reporters are now seeing things?
It’s doing exactly what we had hoped it would do. You can do a Google news search for our names and you’ll get several news stories a week where they’ve used us as sources or are reprinting our work.
We wanted to be part of the conversation, and, to the extent that the things we are saying are interesting to reporters and pundits, we wanted to influence the conversation that’s going on among those who will write this election into history. I feel that’s definitely happened.